Existing treatments for movement impairments are “off-the-rack” rather than “custom-tailored,” but B.J. Fregly, professor of mechanical engineering and bioengineering at Rice, and CPRIT Scholar in Cancer Research, seeks to remedy that situation.
Fregly has received a National Institutes of Health Research Project Grant to develop open-source software for designing individualized treatments for movement impairments using computational modeling and simulation. The four-year, $2.4 million grant is administered by the National Institute of Biomedical Imaging and Bioengineering.
The software will make it easy to create personalized computer models of individual patients and optimize model-predicted post-treatment function for various treatment options.
The personalized models consider the unique movement control, muscular and skeletal characteristics of patients, and optimization makes it possible to identify better treatment solutions than could be found through clinical intuition. Clinical applications of the software include the design of custom-tailored neurorehabilitation protocols for stroke, physical therapy interventions for osteoarthritis and orthopedic surgery plans for bone cancer.
“The goal of most neuromusculoskeletal modeling research is improving the way we treat movement impairments. Though such models have become much more realistic in recent years, they still have not made a positive impact on the design of clinical treatments,” Fregly said.
“This lack of clinical impact is surprising given that comparable computational technology has transformed the design of airplanes, automobiles and other commercial products. However, a computational approach to treatment design requires reverse-engineering a model of a patient’s neuromusculoskeletal system, which is extremely challenging.”
Though some researchers and clinicians remain skeptical about computational treatment design, Fregly believes the time is right to pursue the new approach, which he has already used to design a patient-specific rehabilitation treatment for knee osteoarthritis.
“With recent improvements in computational modeling capabilities,” he said, “our ability to personalize a neuromusculoskeletal model to represent a specific patient, and then use that model to predict an optimal treatment, is no longer science fiction. All we need is the right tool to simplify the model personalization and treatment optimization process.”
In the short term, Fregly suggests identifying a subset of movement impairments that are amenable to computational treatment design. “This way,” he said, “the field could generate some initial clinical ‘wins’ that would help propel it across the threshold of clinical utility and open up a new paradigm for treatment design.”